45 second product explainer avatar: $11.61 on Sume plus with the image

A 45 second product explainer with a Sume avatar costs $11.61 on the default plus tier when you pass a product_image (45 x 0.258) and $11.025 without one (45 x 0.245). The image adds $0.585 over the clip. Standard is $8.73 with the image (45 x 0.194), and max is $26.10 (45 x 0.58).

What the product image does

product_image is optional and must be a fetchable public HTTPS image URL. Omit it for a productless avatar video. When present, it is a reference for the product that appears in the avatar's scene. The docs describe it as an input and do not publish how long the product stays on screen, so confirm by previewing.

If you only need the product as a backdrop, a photo scene (scene.type: "photo" with an image_url) is the other option, and the two serve different purposes.

Fit the script

45 seconds is about 126 words at 2.8 words per second. Split it into scenes with video_inputs: hook, feature, proof, call to action. The total planned duration must stay in the 4 to 60 second window, and a single final video uses one avatar and one shared scene.

Run an Avatar video preview first. For multi-scene input it returns one still per scene, and later stills are pose-anchored continuations of the first, so you see the product placement before the render.

Limits

Speech is English only per the pipeline prompt. Resolution is 720p. If your platform needs a different ratio, set it at the request: 9:16 is the default, and 16:9 and 1:1 are accepted. A 45 second plus clip at $11.61 is a draft-friendly price; for a first pass, use standard at $8.73 and spend on plus or max only after the script is final.

If the cost is too high

Three levers lower the price without changing the story. Drop to standard for the draft, as above. Shorten to 30 seconds: 30 x 0.258 = $7.74 on plus with the image. Or split the explainer into a 20 second spoken part and a 25 second part that you build from stills and the timeline tools, with no avatar on screen. Which fits depends on how much the face adds for your audience.

Whatever you choose, keep the script under the 60 second limit, as Sume refuses longer estimates.

Check any claim in the script against the product before rendering, because a rerender of the whole 45 seconds repeats the full price. A one-word fix still costs $11.61 on plus with the image, so read the script aloud once and have a second person read it.